Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    phrases related to construction and inspection processes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    CodeAttribute
    -0.58
    </em>
    -0.56
     awkwardly
    -0.55
     weirdly
    -0.54
    ThroughAttribute
    -0.53
     famously
    -0.53
     dodgy
    -0.50
    ’
    -0.50
     flew
    -0.48
    mathcal
    -0.48
    POSITIVE LOGITS
     bezeichneter
    0.76
     étoient
    0.69
     auffi
    0.67
     AppColors
    0.65
    foundland
    0.65
     Efq
    0.64
     eenige
    0.63
     pfalz
    0.63
    ciaż
    0.62
    saraba
    0.62
    Act Density 0.244%

    No Known Activations